[Previous] [Next] - [Index] [Thread Index] - [Previous in Thread] [Next in Thread]

Subject: Re: UKNM: Advertising might actually be dead
From: Ben Thompson
Date: Thu, 5 Mar 1998 19:55:49 GMT

Tom Hukins <tomateborcom [dot] com> wrote

>Mat Morrison wrote, in reply to my post:
>>>I don't see why sites which accept advertising can't be more flexible.
>>>Is there a reason why all the ads they display need to be a given size?

>How do we value banners? The answer to this is simple: buyers will be
>willing to pay anything less than their cost per objective. If an
>impression is worth X, then buyers will pay anything less than X. Sellers
>will be willing to accept anything more than the cost of hosting the
>banners. Ideally they will sell at a price which makes them a profit, but
>if this isn't possible they'll sell at the highest price, provided that
>price is less than the cost of hosting the advert. Such costs include ad
>software, invoicing, the devaluation of the site (adverts inconvenience
>visitors who perceive the site less favourably, weaking the site's brand),
>The market for Internet advertising is very close to what economists call a
>perfect market. A large number of buyers and sellers mean individual firms
>do not have enough power to influence the market in any way. Anyone can
>enter the market; the transactional costs of buying and selling advertising
>can be extremely low. Well-designed ad server software allows low-volume,
>low-cost campaigns to be set up easily. Innovative ad networks such as
>SAFE-Audit <URL:http://www.safe-audit.com/> allow even the smallest sites
>to profit from advertising.

Unfortunately it is nothing like a perfect market. A perfect market has
non-existent entry costs and apart from Link Excahnge I cannot think of a
single site that meets that criteria. With a couple of exceptions Banner
advertising income goes to only a few firms. Remove the top 50-100 sites
from the picture (ie the sites clients wish to be associated with) and what
is left are sites clients do not wish to advertise on. The market might be
perfect, but it does not mean we want to shop there.

Also a perfect market requires a standard product. If the product is not
standard (ie a space 468 by 60 pixels) no one will be interested.

>Those who buy advertising can estimate the value of that advertising by
>running low-volume test campaigns. Statistics for click-through and
>impressions can be monitored in real-time. The effects on sales and
>branding take more time to calculate, but they are equally measurable
>whether standard banner sizes are adhered to or not.

See above. Also I would love to see how you value Branding. Most Accountancy
Bodies have being trying to do so for years with all attempts usually
laughed at by the other bodies. 50% of all advertising spending is wasted,
the question is what 50%.

>So, if we can estimate the value of each campaign, why do we need
>standardisation? I believe that if advertisers innovate and break out of
>the standards they will be able to come up with much more compelling
>creative which will do its job more effectively. This will increase the
>value of Internet advertising, benefitting sides of the industry (buyers
>and sellers).
Because Standardisation means cheapness. If I have to redesign a page for
every advert displayed I would not have time to do any work (in fact I would
have to create 60,000+ pages a day by hand). The idea of non standard sizes
is a misnomer that is at first glance interesting but at second glance

>Traditional media advertising relies on buyers arriving at some vague
>notion of the cost per objective and hoping they're right. The Internet
>allows advertisers to test different sites, different creative and
>different targetting.
>Furthermore, sellers can test advertisements too. Whenever they don't enter
>into an exclusivity agreement, they can test different banners and show
>those which pay them the most if payment is made by click-through or sale.
>Statistics can be monitored in real-time by the ad server software. They
>can also respond to their visitors' reactions, so if a 500K Shockwave ad
>drives visitors away they can stop showing it.
>Thus, the most effective adverts are shown, and the least effective aren't.
>There is no need for standard ad sizes because they don't benefit anyone.

You are taking a valid reason for online marketing and attaching an invalid
idea to it. The chief cost in a WWW site is design, layout and if needed,
programming. Standard Ad sizes = single design costs (of course if a client
of yours is willing to pay me the programming and design costs to
incorporate your non-standard banner then that is different. It is, however,
usually called sponsorship).
>>Campaigns like, say the Levi's I-Candy boxes do move away from the
>>traditional model -- but do they move so far as to be effectively
>>sponsorship? Can we really buy the same sort of targeting, frequency
>>capping, audience tracking and adjustment with this sort of campaign as
>>we can with a "standard" banner?
>>Paradoxically, I believe, standardisation allows us much more
>>flexibility in the way we plan and buy our campaigns than your
>>free-and-easy approach.
>Only if ad space buyers and sellers can't be bothered to measure the effect
>of their campaigns. The value of a 468x60 banner can vary so rapidly that
>to take it as being constant is probably a bad idea.

No, the reasons are simple, straightforward and outlined above. On a budget
of $10,000 as a buyer I would want to see 500,000 banners on different sites
not 100,000 banners over 4 sites with associated redesign costs. Apart from
direct sale sites (where it is easier to offer a direct portion of income
derived (ala Amazon's Associate program)) advertisers want perceived value
(ie increase in Brands net worth) and for that it is quantity/quality I am
after. It would have to be a marvellous campaign to justify paying redesign
costs on a site.

The only exception to this is sponsorship and linkage. No one has done
enough in this field i.e. working out associations and providing links
between relevant sites but I'm sure it will come.

Tom, from the way you stand by your argument I can only assume that you were
not around in the early days when the decision on banners had not been made.
Personally, I remember a discussion with Wired over banner sizes in early
1994, the outcome was standard advert size = standard page layout =
straightforward costs, at which point a standard banner size was agreed upon
and stuck with.

Idealism is wonderful, provided it does not impact on practicality.


Ben Thompson

[Previous] [Next] - [Index] [Thread Index] - [Next in Thread] [Previous in Thread]